Use of adaptive and/or customized compression to enhance the efficiency of digital financial data exchanges

• US 8,538,861 B2
• Filed: 10/11/2012
• Issued: 09/17/2013
• Est. Priority Date: 09/15/2009
• Status: Active Grant

• Current Assignee: Auerbach Group LLC
• Original Assignee: Auerbach Group LLC
• Inventors: Auerbach, Andrew
• Primary Examiner(s): Vezeris, James A
• Application Number:
  US13/649,179
• Publication Number:
  US 20130066762A1
• Time in Patent Office: 341 Days
• Field of Search: 705/37, 301/87
• US Class Current: 705/37
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
• CPC Class Codes: H03M 7/40 Conversion to or from variable length codes, e.g. Shannon-Fano code, Huffman code, Morse code

Abstract

The present invention relates to a method for enhancing the efficiency of digitally communicated data exchanges and to a computer system that implements such a method. The invention particularly concerns the use of adaptive custom compression techniques, binary integers (“bits”), massively parallel processing, database optimization techniques and/or calculation optimization techniques to achieve such enhanced efficiency. The invention is applicable to any digitally communicated data exchange, but is particularly applicable to exchanges of financial information such as financial market buy/sell orders, market making, etc.

Claims

1. A computer-implemented method for enhancing the efficiency of digitally communicating a financial message from a first computer to a second computer of an interconnected computer system, which method comprises the steps:

(A) (1) employing said first computer to generate a data compression coding scheme based on order-units present in an unencoded version of said financial message;

(2) employing said computer-generated coding scheme to produce a database of order-units present in said unencoded version of said financial message; and

(3) employing said produced database to form an encoded version of said financial message;

wherein said data compression technique is customized so that said computer-generated coding scheme allocates shorter codes to order-units having a higher p-value in said unencoded financial message and longer codes to order-units having a lower p-value present in said unencoded version of said financial message;

wherein said p-value is a metric of the moving average of the frequency of said order-unit, weighted by its assigned importance, relative to other order-units in said message(s), such that an order-unit having a higher importance-weighted frequency is assigned a higher p-value than an order-unit having a lesser importance-weighted frequency; and

wherein the message length of said encoded version of said financial message is shorter than the message length of said unencoded version of said financial message; and

(B) digitally communicating said encoded version of said financial message from said first computer to said second computer, directly or via one or more intermediate computers.

3. The computer-implemented method of any one of claims 1-2, wherein said data compression technique additionally employs massively parallel processing.

4. The computer-implemented method of any one of claims 1-2, wherein said data compression technique additionally employs a database optimization technique.

5. The computer-implemented method of any one of claims 1-2, wherein said data compression technique additionally employs a calculation optimization technique.

6. The computer-implemented method of any one of claims 1-2, wherein said data compression technique encodes said financial message into an encoded version having at least 80% efficiency.

7. The computer-implemented method of any one of claims 1-2, wherein said data compression technique encodes said financial message into an encoded version having at least 90% efficiency.

8. The computer-implemented method of any one of claims 1-2, wherein said data compression technique encodes said financial message into a message that is at least 50% shorter than a message communicating said unencoded financial message.

9. The computer-implemented method of any one of claims 1-2, wherein said data compression technique encodes said financial message into a message that is at least 5-fold shorter than a message communicating said unencoded financial message.

10. The computer-implemented method of any one of claims 1-2, wherein said data compression technique employs an Arithmetic Algorithm or a Huffman Algorithm.

11. The computer-implemented method of any of claims 1-2, wherein said financial message is an offer for purchase or sale of a financial interest.

12. The computer-implemented method of any of claims 1-2, wherein said method additionally comprises the step:

(C) fulfilling said financial message, wherein said financial message comprises a request for information relating to a financial interest.

13. The computer-implemented method of claim 12, wherein said request for information relating to a financial interest comprises a request for information relating to the value of the financial interest, a request for information relating to news concerning the financial interest, or an agreement to purchase or sell a financial interest, or a confirmation thereof.

2. A computer-implemented method for enhancing the efficiency of digitally communicating a financial message from a first computer to a second computer of an interconnected computer system, which method comprises the steps:

(A) (1) employing said first computer to generate a data compression coding scheme based on order-units present in an unencoded version of said financial message; and

(2) employing said computer-generated coding scheme to produce a database of order-units present in said unencoded version of said financial message; and

(3) employing said produced database to form an encoded version of said financial message;

wherein said data compression technique is adaptive to permit said computer-generated coding scheme to adjust over time to allocate shorter codes to order-units having a higher p-value in said unencoded financial message and longer codes to order-units having a lower p-value for order-units present in both said unencoded version of said financial message and in at least one previously communicated financial message;

wherein said p-value is a metric of the moving average of the frequency of said order-unit, weighted by its assigned importance, relative to other order-units in said message(s), such that an order-unit having a higher importance-weighted frequency is assigned a higher p-value than an order-unit having a lesser importance-weighted frequency; and

wherein the message length of said encoded version of said financial message is shorter than the message length of said unencoded version of said financial message; and

(B) digitally communicating said encoded version of said financial message from said first computer to said second computer, directly or via one or more intermediate computers.

14. A computer system, comprising a first and a second computer, in digital communication with one another, wherein said computer system is specially adapted for enhancing the efficiency of digitally communicating a financial message, and wherein:

(A) said first computer employs a data compression technique to establish a coding scheme and employs said coding scheme to produce an encoded version of the financial message based on order-units present in an unencoded version of said financial message;

(1) wherein said data compression technique is customized so that said coding scheme allocates shorter codes to order-units having a higher p-value in said unencoded financial message and longer codes to order-units having a lower p-value present in said unencoded financial message;

(2) wherein said p-value is a metric of the moving average of the frequency of said order-unit, weighted by its assigned importance, relative to other order-units in said message(s), such that an order-unit having a higher importance-weighted frequency is assigned a higher p-value than an order-unit having a lesser importance-weighted frequency;

wherein the message length of said encoded version of said financial message is shorter than the message length of the unencoded version of said financial message;

and

(B) said second computer contains a stored copy of said database and, upon receiving said encoded version of the financial message, employs said stored database to decode said encoded version of the financial message.

15. A computer system, comprising a first and a second computer, in digital communication with one another, wherein said computer system is specially adapted for enhancing the efficiency of digitally communicating a financial message, and wherein:

(A) said first computer employs a data compression technique to establish a coding scheme and employs said coding scheme to produce an encoded version of the financial message based on order-units present in an unencoded version of said financial message;

(1) wherein said data compression technique is adaptive to permit said coding scheme to adjust over time to allocate shorter codes to order-units having a higher p-value in said unencoded financial message and longer codes to order-units having a lower p-value for order-units present in both said unencoded financial message and in at least one previously communicated financial message;

(2) wherein said p-value is a metric of the moving average of the frequency of said order-unit, weighted by its assigned importance, relative to other order-units in said message(s), such that an order-unit having a higher importance-weighted frequency is assigned a higher p-value than an order-unit having a lesser importance-weighted frequency;

wherein the message length of said encoded version of said financial message is shorter than the message length of the unencoded version of said financial message;

and

(B) said second computer contains a stored copy of said database and, upon receiving said encoded version of the financial message, employs said stored database to decode said encoded version of the financial message.