会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 5. 发明专利
    • MULTIPLIER WITH TAIL DIGIT ARRAY
    • JPH05204610A
    • 1993-08-13
    • JP247192
    • 1992-01-09
    • NEC CORPKATAYAMA AISUKE
    • KATAYAMA AISUKE
    • G06F7/496G06F7/52G06F7/523
    • PURPOSE:To speed up multiplication by dividedly executing carry processing and the direct output of a tail digit array during the processing of a digit product and obtaining a final product. CONSTITUTION:A multiplicand X consisting of N digits (N>=2), i.e., the 0th digit (x[0]) to (N-1)th digit (x[N-1]) expressed by L-digit-advance numbers (L>=2) is multiplied by a multiplier Y consisting of M (M>=2) digits, i.e., the 0th digit (y[0]) to (M-1)th digit (y[M-1]) expressed by L-digit-advance numbers (L>=2) to find out the product Q of the X and Y. In this case, the array product of an X array x[i]0 i N-1 and a Y array y[j] 0 j N-1 is defined as x[i]*y[j]= P[i][j], the order (k) of the value (i+j) is defined and P[i, j] is formed in the ascending order of (k) based upon equations I to III. Then tail digit array are successively calculated without using accumulation by attaining execution in the digit order of digit products by the repeat of unit operation based upon the output of a tail digit array, the carry of a residual number and addition to the succeeding digit product.
    • 8. 发明专利
    • MULTIPLE RULE TYPE HIGH SPEED MULTIPLICATION SYSTEM
    • JPS55162148A
    • 1980-12-17
    • JP6749579
    • 1979-06-01
    • KATAYAMA AISUKE
    • KATAYAMA AISUKE
    • G06F7/53G06F7/493G06F7/508G06F7/52G06F7/523G06F7/72
    • PURPOSE:To obtain the product remainder of a true product by composing the system of two exponent index transformation ROMs, receiving multiplicands X and multipliers Y as addresses, one inverse exponent transformation ROM and a h-digit adder. CONSTITUTION:Multiplicand X and multiplier Y are h-digit binary numerals, which are divided into high-order and low-order digit bits XH and XL, and YH and YL. Then, ROM1 stores the product of new low-order digit split numerals XL and YL and the digits consists of 2 Xh bits. Next, ROMs 2 and 3, of the same constitution, store only low-order bits of the product of high-order and low-order bits of an address input and when the address input of XHYL and XLYH is done, low-order P1L and P2L of the products of high-order and low-order bits are output. To the terminal of h/2-bit adder 4, XLYL=P3H.P3L is applied and the sum bit is applied to the next h/2-bit adder 5 together with output P2L of ROM3 to obtain product remainder rp2 in the 2 modulus as the sum bit. The time of finding the product remainder is the sum of the time of access to ROMs, but since there is no need to consider a carry when the number of digits is small, the addition time can be ignored and high speed multiplication comes into effect.