1. What is a DSP chip DSP chip, also known as a digital signal processor, is a microprocessor with a special structure. The interior of the DSP chip uses the Harvard structure of the program and data, with special hardware multipliers, widely used pipeline operation, providing special DSP instructions, can be used to quickly implement various digital signal processing algorithms. Depending on the requirements of digital signal processing, the DSP chip generally has some main features: (1) a multiplication and primary addition can be completed in one instruction cycle. (2) The program and data space are separated, and instructions and data can be accessed simultaneously. (3) There is a fast RAM in the chip, usually can be simultaneously accessed in two blocks through a separate data bus. (4) Hardware support with low overhead or no overhead cycle and jump. (5) Fast interrupt processing and hardware I / O support. (6) Has a plurality of hardware address generators operating within a single cycle. (7) Multiple operations can be performed in parallel. (8) Support for pipeline operation, allowing an operation such as finger, decoding, and execution to overlap. Other universal functions of the DSP chip are relatively weak compared to the general microprocessor.
2. The world's first single-chip DSP chip in the world of DSP chips is a major milestone in the DSP chip announced by the US $ 1978 AMI, announced by the US IIntel, announced by the 1978 AMI. There is no single-cycle chip that has a modern DSP chip inside the two chips. in 1980. The μPD7720 launched by Japan NEC is the first commercial DSP chip with a multiplier. The first CMOS process produces floating point DSP chips in Japan, which launched a floating point DSP chip in 1982. In 1983, Japan's Fujitsu company's MB8764, its instruction cycle is 120ns, and has a dual internal bus, which has a big leap in the throughput. The first high-performance floating point DSP chip should be AT & T's DSP32 launched in 1984. In so much DSP chip species, the most successful is a series of products for Texas Instruments, referred to as Ti. TI Company Successfully launched the enlighten generation DSP chip TMS32010 and its series of TMS32011, TMS32C10 / C14 / C15 / C16 / C17, etc., and then launched the second generation DSP chip TMS32020, TMS320C25 / C26 / C28, third generation DSP Chip TMS32C30 / C31 / C32, the fourth generation DSP chip TMS32C40 / C44, the fifth generation DSP chip TMS32C50 / C51 / C52 / C53 and a high-performance DSP chip TMS32C80 / C82, a plurality of DSPs. Since 1980, the DSP chip has developed a sharp development, and the application of DSP chips is increasingly wide. From the perspective of computing speed, the Mac (a multiplication and primary addition) time has been reduced from 400ns (such as TMS32010) in the early 1980s to 40ns (such as TMS32C40), and the processing capacity has increased by more than 10 times. The internal key multiplier components in the DSP chip fall from around 40 of the 1980-induced area to 5 or more, and the in-chip RAM adds an order of magnitude above. From the manufacturing process, in 1980, 4μ N-channel MOS process is used, and the submicron CMOS process is generally used. The number of pins of the DSP chip increases from 1980 to more than 200, and the number of pins is increased, meaning the increase in structural flexibility. In addition, the development of the DSP chip is a large decline in the cost, volume, weight and power consumption of the DSP system. 3. DSP chip classification DSP chip can be classified in three ways. (1) According to the infrastructure, this is classified according to the working clock and instruction type of the DSP chip. If the DSP chip can work normally within any frequency of a clock frequency range, in addition to the calculation speed changes, there is no performance drop, such DSP chips are generally called a static DSP chip. If there are two or more DSP chips, their instruction sets and corresponding machine code machine feet are compatible with each other, and such DSP chips are called consistent DSP chips. (2) Press the data format This is classified according to the data format of the DSP chip. The DSP chip operated in the fixed-point format is called a fixed-point DSP chip. Working in floating point format called DSP chips. Different floating point DSP chips are not exactly the same, some DSP chips use custom floating point formats, and some DSP chips use IEEE standard floating point format. (3) According to the purpose, according to the use of the DSP chip, it can be divided into a general-purpose DSP chip and a dedicated DSP chip. General DSP chips are suitable for ordinary DSP applications, such as a series of DSP chips of TI. Specialized DSP chips are designed for specific DSP operations, more suitable for special operations such as digital filtering, convolution, and FFT.
4. The basic structure of the basic structure of the DSP chip has the basic structure of the DSP chip comprising: (1) Harvard structure. The main feature of Harvard structure is to store procedures and data in different storage spaces, namely the program memory and data storage are two mutually independent memory, each memory independently, independently. Compared with the two memories, the program bus and data bus are set in the system, so that the throughput rate of the data is doubled. Since the procedures and memory are in two separate spaces, the finger and execution can completely overlap. (2) Tailing line operation. The pipeline is related to the Harvard structure, and the DSP chip uses the pipeline to reduce the time of the instruction, thereby enhancing the processing power of the processor. The processor can process two to four instructions in parallel, each of which is at different stages of the pipeline. (3) Dedicated hardware multiplier. The faster the multiplication speed, the higher the performance of the DSP processor. Due to a dedicated application multiplier, multiplication can be completed within one instruction cycle. (4) Special DSP instructions. Special DSP Directive DSP chips are special instructions. (5) Fast instruction cycle. Fast instruction cycle Harvard structure, pipeline operation, dedicated hardware multiplier, special DSP instructions plus integrated circuit optimization design allows the DSP chip instruction cycle below 200ns. 5. The characteristic digital signal processing system of the DSP system is based on digital signal processing, so all features with digital processing: (1) Easy to interface. DSP systems and other systems or devices based on modern digital technology are compatible with each other, such a system interface to implement a certain function is much easier than the analog system and these system interfaces. (2) Convenient programming. The programmable DSP chip of the DSP system allows designers to modify and upgrade software in the development process. (3) Good stability. The DSP system is based on digital processing, with a small impact on the ambient temperature and noise, and high reliability. (4) High precision. The 16-digit system can achieve accuracy. (5) It can be repeated. The performance of the analog system is relatively large, and the digital system is basically unaffected, so the digital system is easy to test, debug and mass production. (6) Easy to integrate. Digital components in the DSP system have a high normative normative, which is convenient for large scale integration. 6. Application Modern Digital Signal Processor of the DSP Chip is an IC circuit that performs high-speed digital signal processing. It is just to adapt to multimedia information social needs and grow rapidly. Today, all DSP devices have been quite rich in the city of electronic devices. The DSP devices in the form of large and small package have been widely used in the field of production of various products, and the DSP application field is still expanding and developing rapidly. (1) Digital mobile phone digital mobile phone despite the complicated tricks, it is basically two categories: high-speed mobile phone and low speed mobile phone. Among them, the high-speed mobile phone as the name is a telephone that is used in high-speed mobile, such as a free call in aircraft, ships, and cars. Although digital high-speed movements have passed standards, the European GSM (GLOBALSTEMFORMOBILECOMMUNICATION) standard is generally applied. Since launching digital cellular telephones, there is now widely used in more than 70 countries around the world. A commonly known as GSM standard digital cellular phone, called digital Big Brother, it has an international roaming function, SIMC (SubscriberidentificationModulecard brings users to use Big Brother's convenience.
It is currently extending data communication service capabilities and it is compatible with ISDN system, for example, BT's Cellnets have provided digital data and fax services using GSM, so Toshiba laptops also have digital big brothers. Low-speed mobile phones are essentially. It is a digitized cordless phone, still maintains a sub-mother-of-active structure of analog cordless phone: the sub-machine is also called a mobile phone, which can be a call from the space within the radius of 100 meters. The master is also called the base. Station, can be used as a left-behind phone in the family, can also be hanging on the wall of the store, on the street wire column, widely distributed. Managed by unified switching facilities to achieve seamlesshandon functionality. This type of low-speed mobile phone standard is much. For example, PHS (PersonalHndy-Phonesystelecommunictay, Japan, South Korea, Southeast Asia), PHS (PersonalHandy-PhoneSystem), and Philips and my country United development of DCCT (DigitalChinacordlesstelephone). Among them, especially in the development of low-speed mobile phones in PHS and DECT, my country's DCCT is still in the design phase due to lack of critical DSP technology. Each mobile phone (including high-speed and low speed) is used to use at least 1 DSP, so high-speed digital mobile phones urgently need a large number of DSP devices. (2) Data modem is known, one of the traditional application fields of digital signal processors, is the modem. Today, the modem acts as a link to communicate with the multimedia information processing system, which is increasingly attached. Especially in recent years, the Internet boom, Fang Xingye, ordinary people surpass on InterNET. Using the PC via a modem via a telephone line, a dial-up connection Internet is already the easiest access form. Since Internet users have increased dramatically, 28.8 kbps modems have become a market in the market. In particular, when the active image information is used to call the activity image information on the PC, it is desirable to use a modem with a higher data transfer speed. In order to adapt to this new demand, international standards have been developed internationally (33.6kbps) modems. This means that higher performance DSP devices are required in the high-speed modem. This 33.6 kbps modem (V.34) is designed for transmitting data, and the DSVD modem is developed based on this, which can transmit data and sound. Undoubtedly, this will require a higher functional DSP device. As the high performance modem constantly appears, it seems that low speed modems are like V. 17 (14.4kbps) has no use of martial arts. In fact, it has just been opposite, now information home appliances, such as the PHS master left-behind, and personal FAX integrated products are listed. That is to say, V. 17 (14.4kbps) modem still has a market. Thus, the DSP required in various modems is also a variety of DSPs. (3) Demand for disk / disc controller With the development of multimedia information, various information storage media products come into born, such as disk storage, CD-ROM, and DVD (DigitalveTileDisk) -ROM new products have been listed.
