ITEM 1. BUSINESS

GENERAL

Aware, Inc. (the "Company" or "Aware") was incorporated in Massachusetts in 1986. During its first seven years, the Company was engaged primarily in research, specializing in wavelet mathematics, image and video compression, and channel modulation and coding. The Company holds nineteen patents in areas related to wavelet mathematics, data compression and similar technologies. The Company's revenue during this period consisted largely of research grants from agencies of the U.S. government and certain commercial companies. In 1993, the Company began to shift its business from contract research toward development of: (i) Digital Subscriber Line ("xDSL") technologies, and (ii) image compression products. Two principal lines of business emerged as a result of the decision to commercialize the Company's core technology: telecommunications and image compression.

The Company's telecommunications business is dedicated to developing technology and products that increase the speed of data communications over conventional copper telephone networks. The Company believes that its technology and products will enable telephone companies ("telcos") to utilize their installed bases of copper telephone lines to provide both residential and business customers with interactive data transmission at speeds much higher than currently available. The Company's core telecommunications technology includes algorithms, software, hardware designs, and chipsets that implement Asymmetric Digital Subscriber Line ("ADSL"), splitterless lite Digital Subscriber Line ("DSL Lite"), Very High Speed Digital Subscriber Line ("VDSL"), and Symmetric Digital Subscriber Line ("SDSL") technologies.

The Company has co-developed an ADSL chipset with Analog Devices, Inc. ("ADI"), a leading supplier of integrated circuits. ADI has a non-exclusive technology and software license to manufacture and sell such chipsets for which the Company receives royalty payments. In 1997, the Company entered into an agreement with Lucent Technologies, Inc. ("Lucent") to develop DSL Lite software that will operate on Lucent digital signal processors ("DSPs"). Lucent has a non-exclusive technology and software license to manufacture and sell such chips for which the Company will receive royalty payments. The Company's telecommunications business is also engaged in the design and development of access routers, modems, transceiver modules, and other communications products that incorporate the Company's technologies.

The Company's image compression products include WSQ by Aware, AccuPress for Radiology, AccuPress for Remote Sensing, AccuPress for Multimedia, and SeisPact. In addition, the Company's image compression organization continues to perform some contract research for the U.S. government.

The Company's executive offices are located at 40 Middlesex Turnpike, Bedford, Massachusetts, 01730, and its telephone number is (781) 276-4000.

PRODUCTS AND MARKETS

TELECOMMUNICATIONS

TELECOMMUNICATIONS MARKET

With the rise of the Internet and World Wide Web, network service providers are experiencing a fundamental shift in the type of communications traffic transmitted over their networks. The existing network infrastructure of twisted-pair copper wiring, which was originally designed to provide analog voice service ("Plain Old Telephone Service" or "POTS" ), and fiber coaxial cable, which was designed to provide broadcast cable television service, are increasingly required to carry large amounts of data produced by computers. Service providers are faced with the challenge of providing high-speed data communications at reasonable costs, while preserving their investment in copper wire and coaxial cable networks.

Copper wire telephone networks are estimated to include over 170 million lines in the United States and over 700 million lines worldwide, according to industry sources. These networks represent a massive undepreciated capital investment. Cable television service is currently available to approximately ninety percent of the homes in the U.S. and approximately sixty-five percent of the homes in the U.S. subscribe to the service.

To date, the telcos' copper wire and cable companies' infrastructures have not proven adequate for the increasing volume of traffic generated by computers remotely connected to each other and the Internet. Digital information requires more bandwidth than traditional analog voice communications if it is to be transmitted at a speed that is satisfactory to the computer user. Currently, the fastest transmission rate readily available to typical home or remote office computer users over existing copper wire is achieved through the use of a 56 kilobits per second ("Kbps") modem, although many users still employ modems that are slower than this. For the over 80 million and growing Internet users, these transmission rates are one of the chief frustrations of using the World Wide Web, which is the fastest growing and most data intensive segment of the Internet.

Service providers, recognizing the need for higher speed data communications, are increasingly seeking to upgrade their networks. The telcos are replacing copper wire with fiber optic cable, which permits high speed data transmissions, particularly through the backbone of the network that links their central offices to one another. However, installing fiber optic cable all the way into customers' homes or businesses is prohibitively expensive and would take decades. Similarly, cable companies have deployed hybrid fiber coaxial ("HFC") networks, and are providing two-way data transmissions over these networks using cable modems.

Telcos are seeking cost-effective technologies to accommodate high speed data transmission over copper wires. Some of these technologies are described below:

ISDN. In the early 1980s, telcos introduced Integrated Service Digital Network ("ISDN") technology, which provides digital transmission over copper wire typically at basic rates up to 144 Kbps. Although this technology is several times faster than a voiceband modem, the market penetration of existing ISDN technology is limited because its equipment and installation costs are relatively high, and it does not allow simultaneous POTS and data transmission on those wires.

T-1. T-1 (E-1 in countries outside the U.S.) is a multiplexing format that allows digital conversion of an analog line. Once converted, a T-1 digital line can deliver data at speeds up to 1.544 megabits per second ("Mbps"). However, T-1 service cannot use the existing copper wire networks without expensive and time-consuming modifications, including installation of repeaters every 3,000 to 5,000 feet to regenerate the signal as it passes along the line. T-1 also requires two sets of twisted-pair copper wires and does not allow simultaneous POTS and data transmission on those wires.

HDSL. In 1992, telcos introduced High bit-rate Digital Subscriber Line ("HDSL") technology, which reduces the cost of installing T-1 service. HDSL increases the distance of T-1 transmission over copper wires to approximately 12,000 feet, which reduces the need for repeaters. As a result, some telcos are deploying HDSL technology in their local access networks. However, HDSL requires two sets of twisted-pair copper wires and does not allow simultaneous POTS and data transmission on those wires.

ADSL. For several years telcos have been evaluating the deployment of ADSL technology, which uses digital signal processing technology to expand the useable bandwidth of copper telephone wire. ADSL was initially created in the late 1980s by Bellcore, the research entity jointly created and funded by the Regional Bell Operating Companies ("RBOCs"). ADSL technology allows non-repeated transmissions of data at a distance of up to 18,000 feet over telcos' existing copper networks at a rate of up to 8 Mbps downstream to the customer and at a rate of up to 768 Kbps upstream from the customer, with the speed of transmission decreasing as distance increases. ADSL allows simultaneous POTS and high speed digital data transmission on a single set of twisted-pair copper wires. To accomplish this, ADSL uses a filtering device, called a POTS splitter, to isolate the POTS and ADSL signals from one another.

DSL Lite. In 1997, the Company helped pioneer a new xDSL technology alternatively referred to as DSL Lite, G.lite, splitterless DMT, and Universal ADSL ("UADSL"). Similar to ADSL, DSL Lite enables voice and high-speed data communications to coexist on a single copper telephone wire. DSL Lite has the added advantage that it eliminates, in most cases, the need for a POTS splitter to separate voice and data signals on the customer premise side of the connection. The omission of a POTS splitter vastly reduces the installation cost of DSL Lite service over that of ADSL service by eliminating the need to send a technician for service installation. DSL Lite technology allows downstream data transmissions speeds of up to 1.5 Mbps without any disruption to the customer's telephone service. Also in 1997, a consortium of leading companies from the personal computer, networking and telecommunications industries formed a group known as the Universal ADSL Working Group ("UAWG") to write a specification for UADSL. Upon completion of the specification, the UAWG intends to submit the document to the International Telecommunications Union ("ITU") for consideration as the worldwide G.lite standard.

In addition to these telco technologies, cable company suppliers are working to improve HFC technology, which would permit two-way broadband digital communications over typical cable networks. HFC technology uses digital signal processing to allow efficient sharing of upstream bandwidth so that a cable line can be used for two-way transmissions. According to industry sources, approximately 100,000 lines of cable modems had been installed as of December 31, 1997. New HFC networks are also being installed by telcos so that they can offer television service as well as telephone and data dial-tone services.

Telcos typically put new products through a rigorous approval process before deploying them on a broad basis. The approval process usually involves a number of different phases, including (i) laboratory evaluations, in which the product is tested against relevant industry standards; (ii) technical trials, in which the product is tested in the field with a small number of users; (iii) marketing trials, in which the product is tested in the field with a larger number of users and telcos begin to train their personnel to install and maintain the product; (iv) initial commercial deployment, in which telcos make the product available to selected customers for selected applications; and (v) commercial deployment, in which telcos make the product available to a substantial number of customers.

During 1997, telcos continued laboratory, technical, and marketing trial evaluations of ADSL. While the Company believes the telcos' trial experience has provided them with evidence that ADSL technology is capable of delivering high-speed data transmissions rates, the telcos have been slow to initiate commercial deployment. The Company believes that the telcos have moved cautiously due to two primary reasons: (i) they have not had commercially deployable central office equipment available to them and (ii) they have been concerned about their ability to meet the demand for customer premises equipment installation requests should ADSL become widely available. The Company believes that DSL Lite will act as a catalyst to increase the speed at which the telcos adopt xDSL technologies, including DSL Lite and ADSL. Since DSL Lite largely solves the customer premises installation issue, the telcos and telecommunications equipment suppliers appear to have increased their commitment to deployment as evidenced by the formation of the UAWG. While the timing of wide-scale xDSL deployment is difficult to predict, recent events suggest the industry is increasing the pace at which it is moving toward that objective.

TELECOMMUNICATIONS PRODUCTS

The Company designs and develops products utilizing its proprietary software to implement xDSL that it believes have advantages over its competitors' xDSL products. The xDSL products developed by Aware incorporate proprietary software and algorithms based on digital signal processing technology as well as application specific integrated circuits (ASICs). In contrast to the approach taken by some competing developers of xDSL technology, Aware's approach is to maintain a high level of functionality in the software component of the product as opposed to the ASIC. The Company believes that this approach allows it to engineer improvements in its technology quickly and efficiently, rather than having to design and produce a new ASIC each time an improvement is made. The Company's xDSL technology enables data communications protocols, such as Frame Relay, TCP/IP, and ATM, to operate at higher transmission rates over copper wire. The Company has chosen to use the multi-carrier Discrete Multi-Tone ("DMT") modulation for ADSL, rather than the single-carrier Carrierless Amplitude Phase ("CAP") modulation technique. The Company believes that xDSL/DMT technology has greater potential for deployment than CAP, because (i) DMT is more flexible, (ii) the standardization process for DMT is more advanced, (iii) DMT has been endorsed by the UAWG and the ITU for DSL Lite, and (iv) there are multiple vendors who supply DMT as opposed to CAP which is offered by one vendor. (See Item 1. Business - TECHNOLOGY)

Existing Telecommunications Products

Chipsets. The Company and ADI developed a second generation ADSL chipset, which began shipping in June 1997. The chipset uses a combination of ASICs, digital signal processors, and proprietary software to provide all of the ADSL transceiver functions necessary in a modem chipset. The ADSL chipset meets the performance objectives of the DMT multi-carrier modulation chosen by the American National Standards Institute ("ANSI") as the standard for ADSL.

In 1993 ADI and the Company entered into an agreement, under which ADI produces and markets chipsets incorporating the Company's DMT-based ADSL technology, and for which the Company receives royalties and development funding. Effective March 1998, ADI and the Company modified this agreement from an exclusive to a non-exclusive business relationship. Even though the Company may enter into additional semiconductor partnerships, the Company's ability to achieve its business objectives will still depend on ADI's ability and desire to deliver chipsets to the market place. (See Item 7. Management's Discussion and Analysis of Financial Condition and Results of Operations - FACTORS THAT MAY
AFFECT FUTURE RESULTS.)

Software and Hardware Interfaces. The Company develops software and hardware interfaces for its ADSL chipset which can be used to connect the chipset with PCs, network and central office equipment, and other telephony and data communications devices. The interfaces are custom developed by the Company for OEMs, who seek to incorporate the Company's ADSL technology into their systems.

Access Routers. The Company has developed and markets an ADSL access router, called the x200 Access Router. The x200 contains the ADI/Aware ADSL chipset along with software and hardware interfaces and designs developed by the Company, and routing capability for Frame Relay, TCP/IP, and ATM data communications protocols. In a typical configuration, the Company's ADSL access router is designed to receive data at speeds over 9 Mbps and send data at speeds of up to 768 Kbps, is fully rate adaptive in 32 Kbps increments and is capable of transmitting data over distances of more than 18,000 feet over standard copper wire while maintaining POTS service through the use of a POTS splitter.

Board-Level Products. The Company has developed and markets an ADSL transceiver module, called the AW-910. The AW-910 is a 3" by 5" transceiver card, which contains the ADI/Aware ADSL chipset and software and hardware interfaces. The AW-910, which is primarily sold to OEMs for inclusion in their ADSL product offerings, also transmits and receives data at rates over 9 Mbps downstream and 768 Kbps upstream.

Test and development systems. The Company has developed and markets an ADSL test and development system, called the ADS-910. The ADS-910, which is designed to help OEMs test their ADSL systems, allows a wide range of tests, including ADSL line testing and bit rate testing. The development system houses two transceiver modules; one central office (CO) module and one remote terminal (RT) module, and provides power, data, and ADSL line signals to the modules.

PRODUCTS UNDER DEVELOPMENT

The Company has publicly announced the following product development projects:

Chipsets. The Company and ADI are currently developing their next generation full rate ADSL chipset. This release includes improvements to the two ASICs in the chipset as well as upgrades to the signal processing software for the DSP in the chipset.

The Company has announced that it is working with Lucent to develop DSL Lite software that will operate on Lucent DSPs along with analog front end chips. The resulting chipset will be targeted at personal computer OEM customers. The Company is also working with ADI to develop DSL Lite software that will operate on various ADI chip platforms. The resulting chipset will be primarily targeted at central office OEM customers.

The Company has begun in-house design and development of a chip that implements the Company's proprietary DWMT technology. The resulting chip is intended to have applications in the SDSL marketplace. The chip will be fabricated by a third party semiconductor manufacturer.

Access routers and transceivers. The Company has announced that it intends to begin shipping x200 Lite Access Routers that contain DSL Lite functionality and AW-910 DSL Lite transceiver modules in the first half of 1998. These products, which will incorporate the second generation ADI/Aware ADSL chipset operating in DSL Lite mode, will be rate adaptive and support speeds up to 1.5 Mbps downstream and 512 Kbps upstream and will achieve high-speed data transfer over local loops of up to 22,000 feet.

IMAGE COMPRESSION

In 1993, the Company began an effort to produce commercially marketable wavelet image compression software products. The Company currently offers five software-based compression products and has an agreement with ADI to produce and market a wavelet video compression ASIC, for which the Company receives royalties. The Company's compression products include the following: WSQ by Aware (which compresses digital fingerprint data for use by law enforcement agencies, such as the FBI); AccuPress for Radiology (which is used to compress digital radiographs and other types of medical imagery); AccuPress for Multimedia (which is a general purpose compression product); AccuPress for Remote Sensing (which is designed for compression of satellite-based remote sensing imagery); and SeisPact (which companies in the oil and gas industry can use to store and transmit large amounts of seismic data).

TECHNOLOGY

The Company's core technology is based on its research into wavelet mathematics, digital communications, and data compression. From that core technology, four principal technologies have emerged, including: (i) DMT-based ADSL technology, (ii) splitterless DMT DSL Lite technology, (iii) DWMT technology, and (iv) image compression technology.

ADSL Technology

ADSL is a method for expanding the useable bandwidth of copper wire. Typically, ADSL systems divide a one megahertz (MHz) bandwidth on copper wire into three segments: (i) the 0 to 4 kilohertz (KHz) range is used for POTS, (ii) the 25 KHz to 100 KHz range is used to transmit data upstream and (iii) the 100 KHz to 1 MHz range is used to transmit information downstream. The ANSI specification for ADSL calls for operation rates of 1.5 to 8 Mbps downstream and 64 to 640 Kbps upstream when operating over existing copper wires at a distance of up to 18,000 feet.

There are two primary ADSL modulation techniques for transmitting data signals: (i) DMT, which the Company uses, and (ii) CAP. DMT is a multi-carrier modulation technique that was chosen by ANSI as the telecommunications industry standard for ADSL. CAP is a single-carrier modulation technique originally developed by AT&T Paradyne Corporation (now Globespan Technologies, Inc.). The fundamental difference between CAP and DMT is that CAP treats each of the upstream and downstream frequency ranges as a single element over which as many information bits as possible are transmitted. In contrast, DMT divides the upstream and downstream bands into groups of different smaller subchannel frequency ranges (approximately 4 KHz each) into which a much smaller number of bits are coded and transmitted simultaneously.

The Company believes that DMT technology is better able than CAP technology to address the inherent problems of the telcos' copper wire networks. Because of its multiple small frequency bands, DMT is able to adjust and adapt the information signal to both extract more throughput from a wire and to avoid sending information into frequency ranges that are not useable. Since CAP treats the entire frequency range as a single element, it does not have the ability to balance as easily the use of the frequency spectrum to match efficiently the performance of a given wire.

Splitterless DMT DSL Lite Technology

In 1997, the Company helped to pioneer a new form of xDSL technology, known as splitterless DMT DSL Lite technology. DSL Lite allows for downstream data transmissions over telephone networks in the 1.0 to 1.5 Mbps range, and upstream data transmissions of up to 512 Kbps in a "splitterless" environment. Splitterless means that modems employing this technology do not require special equipment, known as POTS splitters, to separate the telephone service from the data service. The omission of a POTS splitter vastly reduces the installation cost of DSL Lite service over that of ADSL service by removing the need to send a technician for service installation. Further, this technology will enable end users to install DSL Lite modems in the same way as today's voiceband modems. For these reasons, the Company believes that DSL Lite technology may act as a catalyst to increase the speed at which service providers deploy xDSL technologies.

DMT operating in a splitterless DSL Lite mode has been endorsed by both the UAWG and the ITU for two primary reasons: (i) DMT DSL Lite can be made to interoperate with full-rate DMT-based ADSL, the ANSI standard for ADSL modulation, and (ii) DMT DSL Lite equipment and software provide a clear migration path to full rate standards-based ADSL central office and customer premises equipment.

The UAWG intends to submit a specification for splitterless DMT DSL Lite to the ITU in 1998. While predicting the timing or outcome of standard body actions is difficult, the ITU has preliminarily indicated that it will attempt to adopt a standard for this technology by the end of 1998.

DWMT Technology

The Company has invented a proprietary technology based on wavelet mathematics called DWMT. The Company believes that, as a result of its research and development of DWMT technology, it is a leader in commercialization of wavelets for telecommunications.

Multi-carrier systems divide a frequency range into the desired number of subchannels by using a time-domain to frequency-domain transform, which is a mathematical process. Because of fundamental limits associated with such transforms, the process of creating isolated subchannels is imperfect. These imperfections inhibit modems from achieving theoretical performance limits. The subchannelization method used in creating DMT modems utilizes a Fourier transform. This technique has been used in the telecommunications industry since the 1960s, but has become more practical for high speed, high volume use as digital signal processors have improved. The wavelet transform yields significantly better subchannelization than the Fourier transform. Because this technique more closely approximates ideal subchannelization, the performance of a wavelet-based DWMT system can produce performance superior to a non-wavelet DMT system operating in a noisy environment.

The Company intends to apply DWMT technology to new products using SDSL, VDSL, and HFC applications. The Company is seeking to incorporate DWMT techniques into industry standards body

recommendations. The following is a brief description of possible applications using SDSL, VDSL, and HFC:

SDSL. Symmetric Digital Subscriber Line technology is similar to ADSL, but allows two-way data transmission at the same rates. The Company is developing an SDSL application using its DWMT technology. SDSL is not an officially defined standard, but can provide data transmission rates from 1 Mbps to 10 Mbps simultaneously in both directions on single twisted-pair copper wire over distances of 5,000 to 18,000 feet. The Company expects that this SDSL application can be used for LAN interconnecting and enhanced telephony applications.

VDSL. The Company believes that Very high-speed Digital Subscriber Line technology will be the next generation of high-speed user access, critical to the implementation of fiber-to-the-neighborhood and fiber-to-the-curb architectures. These architectures involve the deployment of an access node that utilizes fiber optic cable from a telco's central office to the access node, thus bringing fiber closer to the user. The final connection to the user is new or existing copper wire or new coaxial cable. VDSL is being designed with the objective of providing performance up to six times faster than ADSL, but over a shorter distance. The goal of VDSL is to enable telcos to provide a combination of digital TV, data dial-tone and regular telephony service on a single twisted-pair of copper wire.

HFC. By using the frequency band from 5 to 40 MHz for upstream transmission and the frequency band from 450 to 750 MHz for downstream transmission, it is possible to provide two-way services, such as telephony and data communications, on existing HFC networks. Each of these frequency bands is typically divided into smaller bands, 1 to 2 MHz wide. The Company's HFC technology is based upon DWMT and can provide up to 8 Mbps transmission over a 2 MHz band. HFC telephony and cable modem technology enables cable companies to re-use their existing network to provide two-way data transmission services.

Image Compression Technology

Since 1988, the Company has developed expertise, trade secrets, and intellectual property in the field of wavelet transform-based data compression and has obtained several patents in this area. The Company's wavelet compression technology enables digital image, video and certain types of data to be compressed to between 1% and 10% of their original size. Using wavelet compression, the decompressed data are not bit for bit identical to the original data. A risk with this technique is that, as the original data get smaller, a larger amount of error is introduced into the decompressed data. However, compressed data can be transmitted across networks faster and storage costs are reduced.

RESEARCH AND DEVELOPMENT

The Company believes that its future success depends on its ability to adapt to the rapidly changing telecommunications environment and to meet its customers' needs. The timely development and introduction of new products is essential to maintain the Company's competitive position. The Company's product development activities are focused on delivering technology and products to its OEM customers that will enable them to offer end-to-end systems that will allow their service provider customers to make maximum use of the capabilities of their existing copper wire networks. Key development objectives include enhancements to the Company's ADSL and DSL Lite technologies as well as on products incorporating DWMT technology for SDSL and VDSL applications. In 1996 and 1997, the Company spent approximately $1,100,000 on the development of its HFC technology. In the third quarter of 1997, the Company suspended development of its HFC technology due to a lack of customer demand for such technology at that time. The Company may recommence HFC development activities if and when there is sufficient customer demand.

Most of the Company's products are developed internally. As of December 31, 1997, the Company had a research and development staff of 47 employees, including ten employees holding doctorate degrees related to digital signal processing and digital communications theory. The Company supplemented its staff of 47 engineering employees with 8 additional contract engineers at December 31, 1997. Subject to its ability to effectively source, hire and retain engineers, the Company anticipates that its research and development organization will grow significantly in the future as the Company attempts to strengthen its technology and product position in the telecommunications marketplace.

During the years ended December 31, 1997, 1996 and 1995, research and development costs charged to operations were $6,874,137, $3,234,799, and $2,333,200, respectively. Such costs are net of software development costs capitalized in accordance with Statement of Financial Accounting Standards ("SFAS") No. 86. There were no SFAS 86 costs capitalized in 1997, 1996 or 1995.

New product development schedules are difficult to predict, because telecommunications product development, quality assurance testing and debugging are complex processes that often take longer than expected. Accordingly, although the Company estimates the shipment dates of proposed new products for internal purposes, such estimates are subject to frequent adjustment based on the Company's own periodic assessment of its progress in the development process. No assurance can be given that any of the development projects referred to in the "Products and Markets" section will be successful or that any announced shipping dates for new products will be met.

SALES AND MARKETING

To date, the Company's principal telecommunications sales and marketing strategy has been to propagate its technology and products with OEM equipment suppliers. These OEM customers manufacture and sell telephone network equipment, cable plant equipment, data communications equipment, and end user customer premises equipment. The Company's objective is to incorporate its technology and products into solutions offered by its OEM customers. The Company has three types of OEM customers with whom it has business relationships: (i) licensees of intellectual property, such as U.S. Robotics/3Com, (ii) semiconductor manufacturing partners, such as ADI and Lucent, and (iii) customers who purchase the Company's existing telecommunications products, such as Advanced Fibre Communications, DSC Communications, Ericsson, Siemens, ADC, Phillips, ECI, Sumitomo, and PulseCom.

Due to the complexity of the Company's telecommunications technology, the Company's sales people must have a high degree of technical sophistication in order to market its products effectively. The Company believes that technology selections involving the Company's products are frequently made at senior levels within a prospective customer's organization. Consequently, the Company relies significantly on presentations by senior management to key employees of OEMs.

As of December 31, 1997, the Company had ten people in its telecommunications sales and marketing organization. Although the Company's primary sales and marketing objective is to sell to OEM's, during 1997 the Company began selling its ADSL products to resellers, Internet service providers and competitive local access providers. As xDSL technologies are more broadly adopted, the Company expects to hire additional sales and marketing employees to support the efforts of senior management with its OEM customers and to create awareness for the Company's products and technology with its OEMs' service provider customers.

The Company sells its software-based compression products primarily through OEMs and systems integrators. As of December 31, 1997, there were three people in the Company's compression software sales organization.

The Company has in the past and expects in the future to derive a substantial portion of its revenues from a limited number of customers. There are relatively few OEM equipment suppliers to whom the Company can sell its technology and products. Consequently, the Company's future success will depend to a large extent upon: (i) the timing and size of future purchase orders for the Company's products from these customers, (ii) the financial and operating success of these customers, and (iii) the success of products offered by these customers that use the Company's technology and products. Any attempt by such customers to seek out additional or alternative suppliers or to undertake the internal development and sale of technology and products comparable to those of the Company could have a material adverse effect on the Company's business, financial condition and results of operations. (See Item 7. Management's Discussion and Analysis of Financial Condition and Results of Operations -
FACTORS THAT MAY AFFECT FUTURE RESULTS.)

The Company derived approximately 16%, 13%, and 12% of its total revenue in 1997 from U.S. Robotics/3Com, the United States government and ADI, respectively. The Company derived approximately 22%, 17%, 12%, and 10% of its total revenue in 1996 from DSC Communications Corporation, ADI, the United States government, and Teltrend, Inc., respectively. The Company derived approximately 23%, 18%, 12%, and 10% of its total revenue in 1995 from ADI, General Instrument Corporation, the United States government, and GSS/Array Technology, respectively. Predominately all revenue in 1997, 1996, and 1995 was sold to unaffiliated customers in North America.

MANUFACTURING

The Company's manufacturing capacity is relatively limited and as such it relies on a third party contractor manufacturer to assemble and test substantially all of its xDSL products. The Company's third party manufacturer is located in Canada, and obtains component parts directly from the Company, and from suppliers chosen by the Company or itself. Other than the ADSL chipset, which is available through ADI, the Company believes that other components for its products are available from a number of suppliers. Further, the Company believes that other qualified third party contract manufacturers exist to assemble and test the Company's products in the event its relationship with its current contract manufacturer is not successful. (See Item 7. Management's Discussion and Analysis of Financial Condition and Results of Operations -
FACTORS THAT MAY AFFECT FUTURE RESULTS.)

COMPETITION

The markets for the Company's products are intensely competitive and the Company expects competition to increase in the immediate future, especially in the emerging ADSL and DSL Lite market. The Company intends to compete on the basis of technology, price, the timing of product delivery, product features, quality, reliability, and customer satisfaction. The Company currently competes, or expects to compete in the future, with the following categories of companies:
(i) other vendors of DMT-based ADSL technology, such as Orckit Communications Limited ("Orckit"), Amati Communications Corporation/ Texas Instruments Inc. ("Amati/TI"), PairGain Technologies, Inc. ("PairGain") and Alcatel Network Systems, Inc. ("Alcatel"); (ii) vendors of alternative ADSL technologies, such as Globespan Technologies, Inc. ("Globespan"), which is currently marketing its CAP-based ADSL technology, (iii) OEMs and other systems integrators, such as Ericsson, Northern Telecommunications, Westell, Cisco, 3Com, and Pairgain, and
(iv) Regional Bell Operating Companies ("RBOCs") and other telcos, who are no longer prohibited from manufacturing telecommunications equipment as a result of deregulation

The Company's success will depend on telcos' willingness to invest in broadband digital services based on xDSL technologies. The Company expects that its xDSL technology and products will compete not only with other products that increase the efficiency of digital transmission technologies over copper wire, such as ISDN for Internet access, but also with other broadband transmission technologies, such as HFC, fiber optic cable, digital broadcast satellite and other wireless technologies. The Company believes its current and future broadband products will permit telcos to upgrade their networks in a flexible and cost effective way, but telcos may choose to deploy products using better established technologies to upgrade their networks including fiber optic cable, which many telcos favor. To the extent that telcos choose to install fiber optic cable or other transmission media between central offices and end users, the Company's business, financial condition and results of operations will be materially adversely affected.

The Company believes that, in the ADSL market, its DMT-based products are more flexible and will enjoy greater potential for deployment than products using the CAP technique, which is a non-standard, proprietary, single-source technology. However, CAP-based ADSL products were introduced prior to the Company's products and are more readily available than the Company's products.

To date, there has been only limited commercial deployment of the Company's competitors' DMT-based ADSL products. Therefore, the Company is uncertain how its products will compare with products sold by Alcatel, Amati/TI, Orckit and PairGain, each of whom manufactures DMT-based ADSL products. Each of these competitors have made claims in their sales literature and elsewhere suggesting that their products provide data transmission rates that are equal to or faster than that of the Company's products. However, the Company does not have access to these products, and therefore has no independent means by which it can corroborate these claims.

The markets for the Company's wavelet image compression technology are competitive, and are expected to become increasingly so in the near future. In addition, the Company's WSQ Finger Print Compression product is an implementation of an open standard and is therefore subject to competition.

Many of the Company's competitors and potential competitors, including the RBOCs, Alcatel, and Amati/TI have significantly greater financial, technological, manufacturing, marketing and personnel resources than the Company. There can be no assurance that the Company will be able to compete successfully or that competition will not have a material adverse effect on the Company's business, financial condition and results of operations. (See Item 7. Management's Discussion and Analysis of Financial Condition and Results of
Operations FACTORS THAT MAY AFFECT FUTURE RESULTS.)

INTELLECTUAL PROPERTY

In the field of telecommunications technology, the Company holds eight patents for applying wavelet mathematics to communications systems. The Company has two pending patent applications that pertain to the application of multi-carrier technology to broadband communications. The Company has also filed multiple provisional patents that pertain to its splitterless DSL techniques. The Company also holds six patents for image compression and processing, three patents for video compression, one patent for audio compression, one patent for certain optical applications and one pending patent for seismic data compression.

Although the Company has patented certain aspects of its technology, the Company relies primarily on know-how and trade secrets to protect its intellectual property. The Company attempts to protect its trade secrets and other proprietary information through agreements with its customers, suppliers, employees and consultants, and through security measures. Each of the Company's employees is required to sign a nondisclosure and non-competition agreement. Although the Company intends to protect its rights vigorously, there can be no assurance that these measures will be successful. In addition, the laws of certain countries in which products incorporating the Company's technology may be developed, manufactured or sold may not protect the Company's products and intellectual property rights to the same extent as the laws of the United States.

While the Company's ability to compete may be affected by its ability to protect its intellectual property, the Company believes that, because of the rapid pace of technological change in the telecommunications industry, its technical expertise and ability to introduce new products on a timely basis will be more important in maintaining its competitive position than protection of its existing intellectual property and that patent, trade secret and copyright protections are important but must be supported by other factors such as the expanding knowledge, ability and experience of the Company's personnel, new technology and products, and product enhancements. Although the Company continues to implement protective measures and intends to defend vigorously its intellectual property rights, there can be no assurance that these measures will be successful.

Many participants in the telecommunications industry have an increasing number of patents and have frequently demonstrated a readiness to commence litigation based on allegations of patent and other intellectual property infringement. Third parties may assert exclusive patent, copyright and other intellectual property rights to technologies that are important to the Company. If the Company is found to have infringed any of such patents, the Company could be subject to substantial damages and/or an injunction preventing it from conducting its business, and the Company's business could be materially and adversely affected.

The Company may also receive notices from third parties regarding the pendency of various patent applications which may be pertinent to the design and operation of xDSL telecommunications equipment. Unless and until patents actually issue, there can be no infringement, and the Company has not examined any such patent applications.

Although third parties may offer to license their patents and their patent applications to the Company, there can be no assurance that any license would be available on acceptable terms should the Company choose to pursue such license or be found to infringe such patents. In addition, there can be no assurance that third parties will not assert infringement claims against the Company in the future, that these assertions will not result in protracted and costly litigation, or that the Company would prevail in any such litigation or be able to license any valid patents from third parties on commercially reasonable terms. Further, such litigation, regardless of its outcome, could result in substantial costs to and diversion of effort by the Company. Litigation may also be necessary to enforce the Company's

intellectual property rights. Any infringement claim or other litigation against or by the Company could have a material adverse effect on the Company's business, financial condition and results of operations. (See Item 7. Management's Discussion and Analysis of Financial Condition and Results of
Operations - FACTORS THAT MAY AFFECT FUTURE RESULTS.)

EMPLOYEES

At December 31, 1997, the Company employed 77 people, including 47 in research and development, 15 in sales and marketing, 3 in manufacturing, and 12 in finance, information systems and administration. All of these employees were based in Massachusetts. As necessary, the Company supplements its regular employees with temporary and contract personnel. At December 31, 1997, the Company had engaged 10 temporary and contract personnel primarily working in research and development. The Company believes that its future success will depend in large part on the continued service of its technical and senior management personnel and upon the Company's continuing ability to attract and retain highly qualified technical, sales and marketing, and managerial personnel. Competition for highly qualified personnel is intense, and there can be no assurance that the Company will be able to retain its key managerial and technical employees or that it will be able to attract and retain additional highly qualified personnel in the future. None of the Company's employees is represented by a labor union. The Company considers its employee relations to be good.