J. Ma, K.K. Parhi, and E.F. Deprettere, "Pipelining of Cordic based IIR digital filters", in Proc. IEEE Int. Conf. Acoust., Speech Signal Processing (ICASSP), pp. 643--646, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....(n) 1 2 3 0 1 y(n) r11 ff fl 0 0 0 0 e(n) r12 r13 r14 r24 r23 r22 r33 r34 r44 G 4 p1 p2 p3 Figure 2: Flow graph representation for recursive least squares minimizations implemented using VLSI technology. It is a very powerful technique for pipelining of recursive signal processing algorithms [28 30]. It can transform a sequential recursive algorithm to an equivalent concurrent one which can then be realized using either pipelining or block (parallel) processing or both. The lookahead transformation is not free from its drawback. It accompanied by an increase in the hardware overhead. In the ....

J. Ma, K.K. Parhi, and E.F. Deprettere, "Pipelining of Cordic based IIR digital filters", in Proc. IEEE Int. Conf. Acoust., Speech Signal Processing (ICASSP), pp. 643--646, 1997.

....orthogonal filter [3] have been developed. Recent research has been directed towards designing pipelined and parallel orthogonal IIR filters for high speed and low power applications. The normalized and ODR lattice filters were pipelined in [4] and the cascade orthogonal filter was pipelined in [5]. The algorithms in [1] 3] have been developed This research was supported in parts by NSF under grant number MIP 9258670. using transfer function approach. As an alternative, state space approach based algorithms have been proposed in [6] and [7] The state space approach has several ....

J. Ma, K.K. Parhi, and E.F. Deprettere, "Pipelining of Cordic based IIR digital filters", in Proc. of IEEE Int. Conf. Acoust., Speech Signal Processing, pp. 643--646, 1997.

....which could increase the maximum filter sample rate to a level independent of the filter order. In fine grain pipelining, a novel method based on constraint filter design and polyphase decomposition techniques [2] is proposed which could increase the maximum filter sample rate to any desired level [25]. The proposed architecture for coarse grain pipelining consists of only Givens rotations, and the one for fine grain pipelining consists of only Givens rotations and a few additions. Both architectures can be efficiently realized using Cordic arithmetic based processors [26] This paper is ....

J. Ma, K.K. Parhi, and E.F. Deprettere, "Pipelining of Cordic based IIR digital filters", in Proc. of IEEE Int. Conf. Acoust., Speech, and Signal Processing (ICASSP), pp. 643--646, 1997.

....and overflow oscillations, and stability in spite of parameter quantization. Recently, fine grain pipelined CORDIC based IIR digital filters were developed. These filters can perform the filtering operations at arbitrarily high sample rates at the cost of linear increase in hardware complexity [5][6]. The pipelined filter architectures consist of only Givens rotations and a few additions which are suitable for CORDIC based VLSI implementations [7] The implementation complexity is mainly determined by the complexity of the rotation evaluation or angle computation. Different approaches for ....

....the filter stop band; while the orthogonality of the filter guarantees the low sensitivity in the filter pass band. Therefore, these filters have good numerical properties over the entire frequency band [4] 5] A typical fourth order CORDIC based IIR digital filter architecture is shown in Fig. 1 [6]. Notice that the critical path in Fig. 1 goes forward and backward through the entire filter structure, which contains 7 multiplications and 7 additions. The maximum sample rate is limited by the computation time in the feedback loop. In order to further increase the maximum throughput, Finegrain ....

[Article contains additional citation context not shown here]

J. Ma, K.K. Parhi, and E.F. Deprettere, "Pipelining of Cordic based IIR digital filters," in Proc. of IEEE Int. Conf. Acoust., Speech, and Signal Processing (ICASSP), 1997, pp. 643--646.

....by the computation time in the feedback loop. In order to further increase the maximum throughput, a fully pipelined CORDIC based IIR digital filter architecture has been developed using the constrained filter design and polyphase decomposition technique. 6 An example (adapted from the reference [6]) of a three level pipelined 12th order CORDIC IIR digital filter architecture is shown in Fig. 4 where the angle Phi parameters are given as follows: F (0) z 3 ) Phi = f(1:195; 3:122) 0:124; 1:943) 0:536; 0:073; 0:531; Gamma0:153)g F (1) z 3 ) Phi = f(0:910; 3:060) ....

J. Ma, K.K. Parhi, and E.F. Deprettere, "Pipelining of Cordic based IIR digital filters", in Proc. of IEEE Int. Conf. Acoust., Speech, and Signal Processing (ICASSP), pp. 643--646, 1997.

....of the denominator polynomial of the transfer function (see [3] 4] for detail) These filters have good numerical properties over the entire frequency band and thus are ideal for VLSI implementations. A typical fourth order cascade orthogonal IIR digital filter architecture taken from reference [6] is shown in Fig. 4. From Fig. 4, we see that the sample rates of these filters are limited due to the presence of feedback loops [7] The critical path of the filter goes forward and backward through the entire filter structure, which contains 7 multiply add operations. As the filter order ....

....through the entire filter structure, which contains 7 multiply add operations. As the filter order increases, the number of cascade stages increases, thus the critical path also increases. A pipelining method based on constrained filter design and polyphase decomposition techniques is proposed in [6] to achieve arbitrarily high sample rate (also referred as fine grain pipelining) for CORDIC Z 1 Z 1 sin F sin F sin F sin F F sin F sin z ) P Z 1 Z 1 F sin F sin 02 F cosF cos F sin F sin F cosF cos F cosF cos sin F sin F z ) z ) F sin F sin F cosF cos 02 sin F sin F ....

J. Ma, K.K. Parhi, and E.F. Deprettere, "Pipelining of Cordic based IIR digital filters", in Proc. of IEEE Int. Conf. Acoust., Speech, and Signal Processing (ICASSP), pp. 643--646, 1997.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC