A 1-Mbit CMOS dynamic RAM with a divided bitline matrix architecture

A 1-Mb dynamic RAM has been fabricated using 1.2-/spl mu/m double-level metal CMOS technology. A novel divided bitline matrix architecture allows the conventional double-polysilicon planar memory cell to be used without sacrificing signal-to-noise (S/N) ratio or die efficiency. Optimized for high bandwidth, the device uses static column circuitry and a 256K/spl times/4 organization to achieve data rates >180 Mb/s at worst-case voltage and temperature conditions. The 5.97-mm/spl times/11.4-mm die incorporates a flexible laser blown fuse link redundancy scheme which can repair a wide variety of fabrication defects. Typical row access and cycle times are 85 and 190 ns, respectively, achieving >21-Mb/s bandwidth in the non-optimized row access mode. Although some DC power is dissipated in static circuitry, active power consumption has been kept to 225 mW (45 mA), and standby power consumption has been reduced to 2.5 mW (0.5 mA).     

A 64-kbit dynamic MOS RAM

A 64-kbit dynamic MOS RAM is developed by using 2 /spl mu/m rule VLSI fabrication technology and low power circuit technology. The 2 /spl mu/m rule VLSI fabrication technology is achieved by improving various aspects of the ultraviolet photolithographic, thin-gate oxidation, arsenic ion implantation, and multilevel interconnection processes. Microminiaturization of the device structure has made the voltage requirements for its MOST threshold voltage and DC supply voltages low. A highly sensitive and low power dissipating sense circuit has been developed for the VLSI RAM. A new level-detecting circuit with a logic threshold which is independent of MOST threshold voltage is proposed. A dynamic address-buffer circuit is also shown. The fabricated 64K RAM has 200 ns of access time, 370 ns of minimum cycle time, and 150 mW of power dissipation under typical supply voltage conditions of V/SUB DD/=7 V and V/SUB BB/=-2 V.     

A 16 384-bit dynamic RAM

A 16-kbit dynamic RAM is described which is TTL compatible on all pins, and fits a standard 16-pin package. A single-transistor storage cell is used which occupies 455 /spl mu/m/SUP 2/. The device is fabricated in n-channel two-layer polysilicon gate technology using conventional design rules. The chip size is 145 by 234 mils. A low-power sense amplifier is used for each 64 memory cells. A special refresh mode is possible in which all 256 sense amplifiers are active, and the entire memory can be refreshed in 64 address cycles.     

A dynamic RAM Cell in recrystallized polysilicon

A new, one-transistor, dynamic RAM cell has been fabricated in beam-recrystallized polysilicon. Placing thin oxides both above and below the storage region doubles the storage capacitance. Complete isolation of the storage region by oxides also reduces the susceptibility of the cell to soft errors from collection of charge injected into the substrate by the surrounding elements or by alpha particles. Long storage times are feasible, being limited only by the leakage of the access transistor. A thick oxide under the bit line reduces the bit-line capacitance, further increasing the ratio of storage capacitance to bit-line capacitance.     

A new high-density low-voltage SSIMOS EEPROM cell

An efficient low-voltage EEPROM cell is described which occupies an area of 135 µm²when fabricated with 3-µm CMOS technology. To charge and discharge the floating gate, the device relies on Fowler-Nordheim tunneling of electrons between the floating gate and a narrow window of the device channel region. In addition, the control gate is positioned so as to shield the remaining portion of the floating gate from the substrate. The cell can be programmed in 10 ms with a nominal WRITE voltage of 16 V and an ERASE voltage of 12 V. The WRITE/ERASE endurance of the cell is in excess of 10⁶cycles, and the data retention has been shown to be greater than 10 years at 125°C.     

New dynamic RAM Cell for VLSI memories

A novel high-alpha-particle-immunity and high-density dynamic RAM cell with readout signal gain is proposed. The cell is composed of a MOSFET for charge transfer, a MOS capacitor for charge storage and a junction FET (JFET) with buried channel under the MOS capacitor. The buried channel is dynamically switched according to whether there is charge-storage or not. The cell has extremely small collection efficiency for charges generated by alpha-particles, and allows a large amount of leakage charges due to its peculiar structure. Thus, it can achieve high packing density.     

A three-dimensional folded dynamic RAM in beam-recrystallized polysilicon

A three-dimensional folded one-transistor dynamic RAM circuit consisting of an access transistor in a beam-recrystallized polysilicon layer above a storage capacitor has been fabricated. Large cell capacitance and low transistor leakage are obtained by use of multiple polysilicon layers and by folding the storage capacitor beneath the access transistor. The resulting storage times are longer than 1 min, several orders of magnitude greater than storage times in a previously published nonfolded dynamic RAM in recrystallized polysilicon [1].     

A concurrent built-in self-test architecture based on a self-testing RAM

Manufacturing test is carried-out once to ensure the correct operation of the circuit under test right after fabrication, while testing is carried-out periodically to ensure that the circuit under test continues to operate correctly on the field. The use of offline built-in self-test (BIST) techniques for periodic testing imposes the interruption of the normal operation of the circuit under test. On the other hand, the use of input vector monitoring concurrent BIST techniques for periodic testing provides the capability to perform the test, while the circuit under test continues to operate normally. In this paper, a novel input-vector monitoring concurrent BIST technique for combinational circuits based on a self-testing RAM, termed R-CBIST, is presented. The presented technique compares favorably to the other input vector monitoring concurrent BIST techniques proposed so far with respect to the hardware overhead, and the time required for the concurrent test to be completed (concurrent test latency). R-CBIST can be utilized to test ROM because it results in small hardware overhead, whereas there is no need to stop the ROM normal operation.     

A 256K dynamic RAM with page-nibble mode

A 5-V 256K /spl times/ 1 bit NMOS dynamic RAM with page-nibble mode is designed and fabricated using 2-/spl mu/m design rules and folded bit-line configuration. Molybdenum disilicided polysilicon is used as the second-level gate to reduce the word-line signal delay. A large 98 /spl mu/m/SUP 2/ cell with Hi-C structure stores the signal charge of 210 fC and provides this memory with wide operating margin. The device is immune to voltage bumping and uses laser programmable redundancy. Typical RAS/CAS access times are 80 ns/40 ns. An average operating current of 50 mA with 80 mA peak at 230 ns cycle time and standby current of 2 mA are achieved.     

A high performance sense amplifier for a 5 V dynamic RAM

Increasing dynamic RAM cell density and the use of a single low-voltage power supply have made it mandatory to store the full power supply voltage in the cell and to be able to detect smaller signals reliably with the initial sense amplifier. The authors present a circuit design approach that restores the cell to a full V/SUB DD/ `1' level, preamplifies the initial charge imbalance before sensing by conventional techniques, and is used in the Fairchild 64K design. Design requirements and a detailed analysis of the amplifier are presented along with simulated results, followed by performance data. The circuit analysis shows how the key design parameters should be chosen and the effects of clock timing variations on the performance of the sense amplifier.     

A block-oriented RAM with half-sized DRAM cell and quasi-foldeddata-line architecture

The authors describe a block-oriented random-access memory (BORAM) based on a series-connected cell concept and a quasi-folded data-line architecture. The series-connected cell concept allows a nearly half-sized DRAM cell even when using the same fabrication process as for conventional DRAMs. The low-noise quasi-folded data-line architecture allows the data-line capacitance to be one eighth the conventional value at the minimum, or the number of cells per amplifier to be 64 times the conventional number at the maximum. In addition, this architecture provides a more relaxed layout for the READ/WRITE circuits. The operation of four series-connected cells is observed successfully through a test device which includes a voltage-to-current conversion circuit, a current-mirror amplifier, and a 0.76-μm² crown-shaped stack-capacitor (STC) cell     

A new architecture for the fast Viterbi algorithm

A novel architecture design to speed up the Viterbi algorithm is proposed. By increasing the number of states in the trellis, the serial operation of a traditional add-compare-select unit is transformed into a parallel operation, thus achieving a substantial speed increase. The proposed architecture would increase the speed by 33% at the expense of a fairly modest increase in area, thus becoming an attractive approach in high-speed applications. A simple example is shown to illustrate the proposed algorithm in maximum-likelihood sequence detector. A comparative synthesis is made to compare the proposed architecture with other approaches, and synthesis simulations confirm the projection of the throughput gain. Also, the proposed algorithm is extended to the block-processing architecture, and we show that an additional 50% speedup is achieved.     

An on-line BIST RAM architecture with self-repair capabilities

The emerging field of self-repair computing is expected to have a major impact on deployable systems for space missions and defense applications, where high reliability, availability, and serviceability are needed. In this context, RAM (random access memories) are among the most critical components. This paper proposes a built-in self-repair (BISR) approach for RAM cores. The proposed design, introducing minimal and technology-dependent overheads, can detect and repair a wide range of memory faults including: stuck-at, coupling, and address faults. The test and repair capabilities are used on-line, and are completely transparent to the external user, who can use the memory without any change in the memory-access protocol. Using a fault-injection environment that can emulate the occurrence of faults inside the module, the effectiveness of the proposed architecture in terms of both fault detection and repairing capability was verified. Memories of various sizes have been considered to evaluate the area-overhead introduced by this proposed architecture     

A 1-Mbit CMOS dynamic RAM with design-for test functions

A 1-Mb CMOS DRAM measuring 4.3/spl times/11.7 mm/SUP 2/ (50.32 mm/SUP 2/) has been fabricated using 1.0-/spl mu/m CMOS double-poly single-metal process technology. Both moat and second-level poly are clad to reduce circuit propagation delays. The chip incorporates two modes of 8-bit parallel READ/WRITE, as well as additional functions for test-time reduction. Eight 1-Mb family members can be produced by metal mask selection. The device uses static column circuitry along with two-stage intermediate output buffers to achieve a typical column address access time of 20 ns.     

A 64 Kbit MOS dynamic RAM with novel memory capacitor

A 64 Kbit dynamic RAM is described. The RAM features a novel memory cell using a polysilicon-dielectric-polysilicon (PDP) capacitor. This structure provides performance and density advantages over the conventional approaches. A new sense amplifier configuration is also described in detail. It multiplexes two pairs of bit lines for each sense amplifier. Thus the number of memory cells per bit line is halved. This reduces the length of each bit line, thereby increasing the signal voltage available to the sense amplifier. A compatible dummy cell design is included in the discussion. Using conservative processing (3.5 /spl mu/m device channel length with 700 /spl Aring/ gate oxide thickness) a die size of 3.2 mm/spl times/7.9 mm is achieved. Experimental data are presented in the text.     

A 70 ns high density 64K CMOS dynamic RAM

A 64K /spl times/ 1 CMOS dynamic RAM has been developed in a double-poly n-well CMOS technology with device scaling to the HMOS III level. A p-channel memory array with n-well protection reduced the operating soft error rate to less than one FIT. Periphery complexity is simplified due to CMOS circuits resulting in a size of 30,464 mil/SUP 2/ with a redundancy efficiency of 68%. The RAM has a typical access time of 70 ns and a CMOS standby power of 25 /spl mu/W. In addition, a static column design offers 35-ns data cycle time for high-bandwidth application.     

A theoretical and experimental analysis of the buried-source VMOS dynamic RAM cell

The buried-source dynamic RAM cell combines a VMOS transistor (VMOST) and a buried junction capacitor to make a one-transistor cell (1TC) providing large storage capacitance, long charge retention, and high density. The threshold voltage, breakdown voltage, and weak inversion current for the forward and reverse modes of operation of the VMOST and the junction capacitance are experimentally related to the nonuniform doping profile of the channel. Equations are developed for the VMOST short-channel threshold voltage and storage capacity of the cell. The charge capacity (per unit of cell area) of the buried-source cell is calculated to be 2.5 times that of the conventional 1TC cell. The cell charge retention time was measured at more than 1 s at 100°C, proving operation of the device as a dynamic memory element. The technology is capable of producing an 80-µm²cell using 4-µm minimum features, no cell contacts, and a single level of interconnect.     

A new type of low power read circuit in EEPROM for UHF RFID

A novel low power read circuit without reference in 1 k-bits electrically erasable and programmable (EEPROM) for UHF RFID is designed and implemented in SMIC 0.18 μm EEPROM process. The read power consumption is optimized using a pre-charge sense amplifier. To improve the performance of the read circuit, a self-detect circuit, a read control logic and a feedback scheme are adopted, combined with a special time sequence. For a power supply voltage of 1 V, an average power consumption of 1.6 μA for the read operation of the EEPROM can be achieved when the read clock frequency is 640 kHz. What is more, with a 110 °C temperature change, the read power consumption variation is as low as 12%. The die size of the EEPROM is 0.15 mm², where the read circuit occupies 0.0125 mm².     

A waveband switching architecture and algorithm for dynamic traffic

The proposed multi-granular optical cross-connect (MG-OXC) architecture and heuristic algorithm significantly reduce the number of used ports and hence the cost of an existing network, and also help in reducing the blocking probability of dynamic connection requests.